LazyQC

LazyQC is an open-source Python framework for lazy, observable-driven quantum simulation. You describe programs and request expectation values first; LazyQC plans the whole batch (observable cones, shared preparations, optional ZX rewrites and exact CX cuts under memory limits) and only runs the native Qulacs simulation when you call collect().

Installation

Requires Python 3.12 or newer. With uv:

uv add lazyqc

Optional extras: zx (PyZX rewrites), cutting (exact CX cuts, Qiskit frontend), tensor and mqt (alternative backends for benchmarking).

Quick start

from lazyqc import Program, plan_all

program = Program(4).ry(0, 0.4).cx(0, 1).ry(2, 0.8).cx(2, 3)
requests = [
    program.expectation({0: "Z"}),
    program.expectation({0: "X", 1: "X"}),
    program.expectation({2: "Z", 3: "Z"}),
]

plan = plan_all(requests, strategy="auto")  # classical planning only
print(plan.explain())
print(plan.collect().values)  # native simulation happens here

More examples live in examples/ and notebooks/, and design notes and results in docs/.

Development

uv sync --locked --all-extras
uv run pytest -q

Contributing

Contributions are welcome. Please read our Code of Conduct before opening an issue or pull request.

Citation

If you use LazyQC in your research, please cite the accompanying paper (see also CITATION.cff):

@article{montalban2026lazyqc,
  title   = {LazyQC: Lazy Planning for Batched Quantum Result Queries},
  author  = {Montalb{\'a}n, Iraitz},
  journal = {TODO},
  year    = {2026},
  doi     = {TODO}
}